INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.2.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

conom

-0.68

iland

-0.67

asket

-0.66

tern

-0.65

inar

-0.61

iban

-0.60

 Argument

-0.59

imony

-0.59

 advisory

-0.59

tsy

-0.58

POSITIVE LOGITS

 respectively

0.92

 albeit

0.83

aka

0.76

ItemTracker

0.74

 namely

0.74

esses

0.72

etc

0.72

 whereas

0.72

 culminating

0.70

 Including

0.69

Activations Density 0.000%

No Known Activations

This feature has no known activations.